The invention lies in the field of detection, characterization and/or elimination of suspended particles in a carrier gas. In particular, it relates to a method and to a device for characterizing, separating and/or eliminating suspended particles in a carrier gas.
Waste gas from emission sources in the household, private transport, goods transport and industry contain suspended particles. A substance mixture of suspended particles and a carrier gas is often called an aerosol. Submicroscopic suspended particles are of particular concern to the public since they may have access to the lungs and may affect the health.
For characterizing emission sources and measurements of suspended particles in waste gases, in a first step an unadulterated sample must be obtained whose suspended particle concentration permits conclusions to be drawn on the quantity of the harmful suspended particles emitted by the emission source. Furthermore, a measuring method should be available which permits an as convincing as possible characterization of the suspended particles whilst taking their noxiousness into account.
Traditional gravimetric methods are used for the particle measurement. The mass of particles for example filtered by or in a filter is measured and serves as an indicator for the noxiousness of a waste gas. The disadvantage is the fact that the small and middle-sized particles are attributed a low weight by this method, and it is those small and middle-sized particles which are particularly harmful due to their access to the lung in contrast to the larger particles.
With the measurement of particle emissions of motors and vehicles, the number instead of the total weight of the emitted particles is therefore significant as a characteristic quantity.
It would be desirable to have at one's disposal a measurement method and a corresponding device which provide measurement results which are more balanced with regard to noxiousness than gravimetric measurements and which for example determine the particle number instead of their total mass. This is also demanded with respect to the creation of new waste gas standards (for example with the projects sketch Particulate Measurement Program (PMP) of UNECE/GPPE). A measurement device should furthermore be as inexpensive and as compact as possible so that it may be applied in measurement test beds of local traffic offices and automobile garages etc.
The invention proceeds along the path of a measurement method and device which fulfils these conditions.
Measurement methods for determining the particle number are known per se. However, the follow problems arise:                1. Dilution: One may never measure the complete flow of waste gas, but a sample should be able to be taken whose suspended particle concentration permits conclusions to be drawn with regard to the total emitted suspended particle quantity. According to the state of the art, a so-called full flow diluter is often arranged upstream of the measurement system. All waste gas is supplied to the measurement line. Additionally, so much dilution gas is admixed that the volume flow in the measurement line is constant independently of the waste gas quantity. In this manner it is achieved that the concentration of the suspended particles in the measurement line is proportional to the emitted total particle quantity. For measurement, any magnitude of part quantity needs to be diverged from the gas in the measurement line and the particle concentration contained therein determined. The measurement systems based on full flow dilution are however very complicated, large and expensive since a very large gas volume needs to be dealt with (the volume flow in the measurement line indeed must be larger than the maximal waste gas flow, e.g. at full load of the motor).        2. Volatile suspended particles: With the suspended particles it is the case of volatile and solid substances. Such a situation is present with diesel exhaust. With diesel exhaust, the solid dust particles consist mainly of carbon, the volatile particles of condensed hydrocarbons and/or sulphuric acid and water. In many cases, it is desirable to separate the volatile from the solid suspended particles or to eliminate the volatile suspended particles. For example, when considering respiratory diseases, it is chiefly the solid suspended particles that contribute to the noxiousness of the waste gas. Furthermore, with regard to the noxiousness of volatile suspended particles, the extent of their noxiousness is the subject-matter of current trials and examinations,—other measurement criteria are significant. Their number to the first degree is not relevant but their total mass.        
For eliminating the volatile suspended particles form the measurement gas there are a few starting points known from the state of the art:
Thermodesorber: The thermodesorber (often also called as thermodenuder) is an apparatus known for several years and commercialized by various suppliers (TSI/Topas; Dekati), for removing volatile particles from an aerosol. The thermodesorber consists of a heated tube in which the aerosol is brought to a defined temperature and thus volatile aerosol components may be evaporated. A so-called activated carbon trap is connected to this, in which the aerosol vapor mixture is led through a tube with wire grating walls on the outside of which an activated carbon (granulated material) is located. Vapor molecules of the volatile aerosol components on account of their strong diffusion movement advance up to the activated carbon granules where they are absorbed and thus removed from the aerosol. The solid, non-evaporated aerosol particles follow the gas flow through the tube with the grating walls. Behind the activated carbon trap, the aerosol on account of the absorption by the activated carbon, contains only such a small amount of vapors that the vapors condense to no or to an insignificant extent. The removal of volatile aerosol components under unfavorable conditions depends on the saturation degree of the activated carbon.
Ejector-diluters (Dekati): The Finnish company Dekai supplies so-called ejector diluters with which the raw aerosol is led through a nozzle and entrained by a rapid dilution air flow. The disadvantages of the ejector method are the narrow setting range of the dilution conditions and its dependence on the pressure conditions.
Measurement difficulties with small particles: Common measurement methods have their limits when counting very small suspended particles which may be significant which regard to their number.
One method for counting suspended particles which is common per se is based on the light scatter of these particles on passage through a light beam. For this, the particles are blown through the continuous light beam by way of a nozzle. By way of suitably dimensioning the nozzle diameter, particle concentration and light beam diameter, one may ensure that, in each case, only a single particle is located in the light beam. With its passage through the light beam, a particle scatters the light in spatial directions outside the beam direction. Photodiodes, which are arranged in these spatial directions, thus detect the rise in the light intensity during the passage of the particle. The number of the thus measured scatter light pulses is thus equal to the number of particles which are jetted into the counting optics.
The intensity of the scattered light is heavily dependent on the size of the particles. Below a particle size of 200-300 nm a detection of the scatter light signal is very complex and practically no longer possible. For this reason, the method of particle counting via light scattering is often applied in combination with so-called condensation nucleus counters (CNC). Since the particles in the submicron region escape the counting with optical methods on account of the low intensity of the light scattered by them, they are “enlarged” in the condensation nucleus counter until they provide a scatter light signal which is sufficient for a counting. For this reason, the aerosol containing submicron particles is firstly led over the saturated vapor of the fluid. Subsequently, by way of strong cooling in a condenser, a supersaturation of the vapor is produced, which as a result of this condenses on the available surfaces, amongst others on the particles. The diameter to which the droplets condense on the particles at the end of the condensation procedure, is largely dependent on the original size of the aerosol particles.
Butanol is often used as an evaporated-on fluid. The particle-containing gas firstly flows through a heated path (35° C.) over a butanol bath. At the same time, the gas is saturated with butanol vapor. Subsequently, a cooled part (condenser, 10° C.) follows, where the vapor is supersaturated due to the reduced temperature and condenses on the particles. The particles thus grow to droplets of typically about 10 μm diameter.
The droplet formation, however, only functions above a certain minimum diameter of the introduced aerosol particles. The supersaturation may only be so large that no homogenous nucleation, that is to say droplet formation without condensation nucleus, occurs. On the other hand, the supersaturation should be as large as possible so that also very small particles may grow (vapor pressure increase with small particles). With a given supersaturation the minimal diameter D is determined by the Kelvin-Gibbs equation:Dk=4σM/(RTρ In(S))
(Dk: Kelvin-diameter
σ: surface tension
M: molar mass of the vapour molecules
ρ: density of the fluid
R: gas constant
S: supersaturation).
The equation applies to droplets of the material of the vapor, thus only if at least one monolayer of vapor is already adsorbed on the particle. The condition that this monolayer arises is dependent on the particle surface.
For the condensation nucleus counter (CNC) this means:                1. The conditions for forming the first layer must be fulfilled (dependent on the particle material, surface structure).        2. The critical supersaturation must be achieved or D>Dk so that the particle may continue to grow.        
Apart from these phenomena, diffusion losses in the CNC may yet further disturb the measurement of very small particles.
Typical lower limits for the particle size lie at 20 nm or even smaller. Today, there are already models which still have 50% efficiency at 3 nm. The end size of the droplet arising in the CNC is largely dependent on the particles size.
With the measurement of particles from motor or vehicle emissions, it may occur that, e.g. due to the formation of condensation products on taking the sample, a not insignificant share of the particles are located in the vicinity of this diameter-related sensitivity threshold of the condensation nucleus counter. The measurement result then depends greatly on the momentary and usually non-stable size distribution of the particles and is, thus, no longer reliable.